We have developed a new algorithm for noise reduction in cardiac transmission scans which involves remapping transmission image values after sharpening the image histogram to account for noise. This method has been tested extensively in phantoms with initial evaluation in patients and found to produce an effective reduction in noise equivalent to lengthening the transmission scan by a factor or two without bias. The project is currently testing the method more thoroughly in patients and refining and optimizing the technique.